Project Abstract Amyotrophic lateral sclerosis (ALS) is a devastating progressive neurodegenerative disease in which the loss of upper (primary motor cortex, M1) and lower (spinal cord, SC) motor neurons (MNs) ultimately leads to total paralysis. MN loss in ALS involves cell autonomous and non-cell autonomous activities in multiple cell types of the M1 and SC, the organization of which are well understood. However, there remain 4 major gaps in our knowledge: 1) How ALS-associated molecular pathology in the various cell types of the M1 relates to those in the SC; 2) How subpopulations of specific cell types are spatially arranged in these two regions; 3) How subpopulations of different cell types are organized in higher-order ensembles; and 4) How the coordinated behavior of these ensembles relates to disease-associated molecular pathology (e.g., pathognomonic inclusions). Towards addressing these questions, we propose to develop a spatially resolved multi-omics catalog of cellular subpopulations in the M1 and SC of patients with ALS and healthy controls. By using a combination of approaches to simultaneously map the spatial transcriptome and proteome of all interacting cellular subpopulations in these regions, our aim is to elucidate the origins and temporal dynamics of inter- and intra-cellular activities that may reveal novel diagnostic and therapeutic targets for ALS. Our overarching hypothesis is that ALS pathology stems from dysfunctional MN-glial interactions, and that this predictably differs in the M1 and SC in accordance with patient symptomatology. To address this hypothesis, we propose to use spatially resolved transcriptomic and proteomic measurements to study intact human postmortem tissue from patients stratified by clinical presentation (i.e., site of initial symptom presentation, bulbar or lower limb). We have previously implemented Spatial Transcriptomics on mouse and human SC to identify regional differences within subpopulations of various cell types that vary as a function of disease dynamics. Here, we propose to build upon our existing human study, and for the first time, develop a spatially resolved multi-omics dataset at scale and in the context of disease in matched human postmortem M1 and SC samples (Aim 1), to enable simultaneous exploration of upper and lower motor neurons in the context of intact tissue. These data will be directly tied to measures of ALS pathology (e.g., pathognomonic inclusions). To integrate and analyze relationships between data across modalities, we will develop a computational framework for harmonized analysis of multi-modal, multi- omic measures of ALS disease burden (Aim 2). Finally, we will implement highly multiplexed immuno-imaging to validate top gene candidates generated in Aim 1 at a single-cell level in situ (Aim 3). We expect to obtain an unmatched view of cellular interactions in the postmortem ALS M1 and SC, and to be able to directly link such interactions to features of ALS pathology in situ. This will allow us to identify dysregulated signaling that drives upper and lower motor neuron loss and associated symptoms in patients in ALS.